Please wait a minute...
浙江大学学报(工学版)
计算机技术、控制技术     
基于最大共识的模型组合算法
董立岩, 朱琪, 李永丽
1. 吉林大学 计算机科学与技术学院,吉林 长春 130012
2. 东北师范大学 计算机科学与信息技术学院,吉林 长春 130117
Model combination algorithm based on consensus maximization
DONG Li yan, ZHU Qi, LI Yong li
1. College of Computer Science and Technology, Jilin University, Changchun 130012, China;
2. School of Computer Science and Technology, Northeast Normal University, Changchun 130117,China
 全文: PDF(1045 KB)   HTML
摘要:

针对原有的随机森林算法没有区别各个单分类器之间的分类优势,对分类器的组合方案进行优化,提出一种基于最大共识的模型组合算法.该算法将分类器的经验误差和泛化误差融入到分类器的权重计算中,充分发挥了单分类器的个性与优势,强化分类效果好的单分类器的优势,弱化分类效果较差的单分类器的劣势.实验结果表明,基于最大共识模型组合算法能够提升组合分类器的分类性能,在提高分类精度的同时,也具有较强的泛化能力,这一改进对于提升同类型多模型组合算法的性能具有一定指导意义.

Abstract:
The combinatorial scheme was optimized and model combination algorithm based on the consensus maximization was proposed, aiming at the problem that the original random forest algorithm can not distinguish the classification advantage between each single classifier. The new algorithm integrated the empirical error and generalization error of the classifier into the classifier weight calculation, which makes each single classifier give full play to their personality and advantages. As a result, this method strengthened the advantage of good classifiers and weakened the disadvantage of poor classifiers. Experimental results show that this optimized algorithm can not only improve the performance of combinatorial classifiers, but also improve the classification accuracy and the generalization ability. This improvement is instructive to improve the performance of the same type of multi-model combination algorithm.
出版日期: 2017-03-06
CLC:  TP 301.6  
基金资助:

国家自然科学基金资助项目(61272209)

作者简介: 董立岩(1966—),男,教授,从事数据挖掘等研究.ORCID: 0000-0001-7491-5893. E-mail: dongly@jlu.edu.cn
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章  

引用本文:

董立岩, 朱琪, 李永丽. 基于最大共识的模型组合算法[J]. 浙江大学学报(工学版), 10.3785/j.issn.1008-973X.2017.02.026.

DONG Li yan, ZHU Qi, LI Yong li. Model combination algorithm based on consensus maximization. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 10.3785/j.issn.1008-973X.2017.02.026.

[1] GALL J, LEMPITSKY V. Decision forests for computer vision and medical image analysis[M]. London :Springer, 2013: 143-157.
[2] GRAY K R, ALJABAR P, HECKEMANN R A, et al.Random forest-based similarity measures for multi-modal classification of Alzheimer’s disease [J]. NeuroImage, 2013, 65: 167-175.
[3] ZHAI S, XIA T, WANG S. A multi-class boosting method with direct optimization [C]∥Proceedings of the 20th ACM SIGKDD international conference on Knowledge Discovery and Data Mining. New York: ACM, 2014: 273-282.
[4] ZHAI S, XIA T, TAN M, et al. Direct 0-1 loss minimization and margin maximization with boosting[C]∥Advances in Neural Information Processing Systems. Nevada: \[s.n.\] 2013: 872-880.
[5] DONG L, LI X, LANG P. Prediction of rockburst classification using Random Forest [J]. Transactions of Nonferrous Metals Society of China, 2013, 23(2):472-477.
[6] HANSEN L K, SALAMON P. Neural network ensembles [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 1990, 12(10): 993-1001.
[7] BAUMANN F, CHEN J, VOGT K, et al. Improved threshold selection by using calibrated probabilities for random forest classifiers [C]∥Computer and Robot Vision (CRV), 2015 12th Conference on. Halifax: IEEE,2015: 155-160.
[8] WANG P, JIANG T, FAN G, et al. Prediction of torpedo Initial velocity based on random forests regression[C]∥Intelligent Human-Machine Systems and Cybernetics (IHMSC), 2015 7th International Conference on. Hangzhou: IEEE, 2015: 337-339.
[9] KREMIC E, SUBASI A. Performance of Random Forest and SVM in Face Recognition [J]. The International Arab Journal of Information Technology, 2016, 13(2): 287-293.
[10]MARATEA A, PETROSINO A, MANZO M. Adjusted F-measure and kernel scaling for imbalanced data learning [J]. Information Sciences, 2014, 257(2):331-341.
[1] 鲁建厦,翟文倩,李嘉丰,易文超,汤洪涛. 基于改进混合蛙跳算法的多约束车辆路径优化[J]. 浙江大学学报(工学版), 2021, 55(2): 259-270.
[2] 付晓峰,牛力,胡卓群,李建军,吴卿. 基于过渡帧概念训练的微表情检测深度网络[J]. 浙江大学学报(工学版), 2020, 54(11): 2128-2137.
[3] 苗峰,谢安桓,王富安,喻峰,周华. 多阶段可替换分组并行机调度问题的求解[J]. 浙江大学学报(工学版), 2015, 49(5): 866-872.
[4] 倪广翼, 章孝灿, 苏程, 俞伟斌. 基于多染色体演化的自适应类别数聚类方法[J]. 浙江大学学报(工学版), 2014, 48(6): 980-986.
[5] 孔勇奇, 潘志庚. 基于抽风矢量场的深度凹陷图像分割算法[J]. 浙江大学学报(工学版), 2014, 48(6): 1024-1033.
[6] 刘加海,杨茂林,雷航,廖勇. 共享资源约束下多核实时任务分配算法[J]. J4, 2014, 48(1): 113-117.
[7] 赵诗奎, 方水良, 顾新建. 柔性车间调度的新型初始机制遗传算法[J]. J4, 2013, 47(6): 1022-1030.
[8] 张俊超, 岳茂雄, 刘华锋. 结构先验约束的动态PET图像重建[J]. J4, 2012, 46(6): 961-966.
[9] 刘毅,李平,高增梁. 用于高炉铁水质量预报的改进支持向量回归[J]. J4, 2012, 46(5): 830-836.
[10] 方水良, 姚嫣菲, 赵诗奎. 柔性车间调度的改进遗传算法[J]. J4, 2012, 46(4): 629-635.
[11] 喻海清, 刘毅, 陈坤, 纪俊, 李平. 鲁棒的递推核学习建模方法在高炉过程的应用[J]. J4, 2012, 46(4): 705-711.
[12] 刘加海,杨茂林. 基于多核处理器平台的公平调度算法[J]. J4, 2011, 45(9): 1566-1570.
[13] 徐敬华, 张树有, 伊国栋, 屠立, 光耀. 面向目标变异的操作臂运动学优化设计[J]. J4, 2011, 45(2): 209-216.
[14] 倪何, 程刚, 孙丰瑞. 基于混合演化的自适应建模及其应用[J]. J4, 2010, 44(8): 1490-1495.
[15] 戴文战, 熊伟, 杨爱萍. 基于函数cot (xα)变换及背景值优化的灰色建模[J]. J4, 2010, 44(7): 1368-1372.